Development and validation of a prognostic signature for preoperative prediction of overall survival in gastric cancer patients

Improve individual treatment by comparing treatment benefits: cancer artificial intelligence survival analysis system for cervical carcinoma

Purpose

The current study aimed to construct a novel cancer artificial intelligence survival analysis system for predicting the individual mortality risk curves for cervical carcinoma patients receiving different treatments.

Methods

Study dataset (n = 14,946) was downloaded from Surveillance Epidemiology and End Results database. Accelerated failure time algorithm, multi-task logistic regression algorithm, and Cox proportional hazard regression algorithm were used to develop prognostic models for cancer specific survival of cervical carcinoma patients.

Results

Multivariate Cox regression identified stage, PM, chemotherapy, Age, PT, and radiation_surgery as independent influence factors for cervical carcinoma patients. The concordance indexes of Cox model were 0.860, 0.849, and 0.848 for 12-month, 36-month, and 60-month in model dataset, whereas it were 0.881, 0.845, and 0.841 in validation dataset. The concordance indexes of accelerated failure time model were 0.861, 0.852, and 0.851 for 12-month, 36-month, and 60-month in model dataset, whereas it were 0.882, 0.847, and 0.846 in validation dataset. The concordance indexes of multi-task logistic regression model were 0.860, 0.863, and 0.861 for 12-month, 36-month, and 60-month in model dataset, whereas it were 0.880, 0.860, and 0.861 in validation dataset. Brier score indicated that these three prognostic models have good diagnostic accuracy for cervical carcinoma patients. The current research lacked independent external validation study.

Conclusion

The current study developed a novel cancer artificial intelligence survival analysis system to provide individual mortality risk predictive curves for cervical carcinoma patients based on three different artificial intelligence algorithms. Cancer artificial intelligence survival analysis system could provide mortality percentage at specific time points and explore the actual treatment benefits under different treatments in four stages, which could help patient determine the best individualized treatment. Cancer artificial intelligence survival analysis system was available at: https://zhangzhiqiao15.shinyapps.io/Tumor_Artificial_Intelligence_Survival_Analysis_System/.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12967-022-03491-8.

Artificial intelligence predictive system of individual survival rate for lung adenocarcinoma

Background

The current research aimed to develop an artificial intelligence predictive system for individual survival rate of lung adenocarcinoma (LUAD).

Methods

Independent risk variables were identified by multivariate Cox regression. Artificial intelligence predictive system was constructed using three different data mining algorithms.

Results

Stage, PM, chemotherapy, PN, age, PT, sex, and radiation_surgery were determined as risk factors for LUAD patients. For 12-month survival rate in model cohort, concordance indexes of RFS, MTLR, and Cox models were 0.852, 0.821, and 0.835, respectively. For 36-month survival rate in model cohort, concordance indexes of RFS, MTLR, and Cox models were 0.901, 0.864, and 0.862, respectively. For 60-month survival rate in model cohort, concordance indexes of RFS, MTLR, and Cox models were 0.899, 0.874, and 0.866, respectively. The concordance indexes in validation dataset were similar to those in model dataset.

Conclusions

The current study designed an individualized survival predictive system, which could provide individual survival curves using three different artificial intelligence algorithms. This artificial intelligence predictive system could directly convey treatment benefits by comparing individual mortality risk curves under different treatments. This artificial intelligence predictive tool is available at https://zhangzhiqiao11.shinyapps.io/Artificial_Intelligence_Survival_Prediction_System_AI_E1001/.

Bioinformatics analysis reveals immune prognostic markers for overall survival of colorectal cancer patients: a novel machine learning survival predictive system

Objectives

Immune microenvironment was closely related to the occurrence and progression of colorectal cancer (CRC). The objective of the current research was to develop and verify a Machine learning survival predictive system for CRC based on immune gene expression data and machine learning algorithms.

Methods

The current study performed differentially expressed analyses between normal tissues and tumor tissues. Univariate Cox regression was used to screen prognostic markers for CRC. Prognostic immune genes and transcription factors were used to construct an immune-related regulatory network. Three machine learning algorithms were used to create an Machine learning survival predictive system for CRC. Concordance indexes, calibration curves, and Brier scores were used to evaluate the performance of prognostic model.

Results

Twenty immune genes (BCL2L12, FKBP10, XKRX, WFS1, TESC, CCR7, SPACA3, LY6G6C, L1CAM, OSM, EXTL1, LY6D, FCRL5, MYEOV, FOXD1, REG3G, HAPLN1, MAOB, TNFSF11, and AMIGO3) were recognized as independent risk factors for CRC. A prognostic nomogram was developed based on the previous immune genes. Concordance indexes were 0.852, 0.778, and 0.818 for 1-, 3- and 5-year survival. This prognostic model could discriminate high risk patients with poor prognosis from low risk patients with favorable prognosis.

Conclusions

The current study identified twenty prognostic immune genes for CRC patients and constructed an immune-related regulatory network. Based on three machine learning algorithms, the current research provided three individual mortality predictive curves. The Machine learning survival predictive system was available at: https://zhangzhiqiao8.shinyapps.io/Artificial_Intelligence_Survival_Prediction_for_CRC_B1005_1/, which was valuable for individualized treatment decision before surgery.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-022-04657-3.

A Modified T-stage Classification for Gastric Neuroendocrine Tumors

BACKGROUND

The eighth edition of the American Joint Committee on Cancer (AJCC) staging manual's TNM staging classification for gastric neuroendocrine tumors has been shown to have poor prognostic discriminability. The aim of present study was to propose a modified T-stage classification, and externally validate its performance in a separate population data registry.

METHODS

A modified T-stage classification with tumor size and extent of tumor invasion was generated from the National Cancer Database between 2004 and 2014 (n = 1249). External validation was performed using the Surveillance, Epidemiology, and End Results registry between 1973 and 2013 (n = 539).

RESULTS

In the National Cancer Database population, using the AJCC T-stage classification, the 5-y survival rates were 85.7%, 80.8%, 64.5%, and 46.1% in T1, T2, T3, and T4 patients respectively (P < 0.001). These rates were more contrasting with the modified T-stage (mT) classification at 87.0%, 78.2%, 59.0%, and 40.3% respectively (P < 0.001). When patients within each of the AJCC T stages were stratified by mT stages, significant survival heterogeneity was observed within each of the AJCC T2 to T4 stages (P < 0.01). Conversely, when mT stages were stratified by AJCC T stage, no survival difference was observed in any of the mT stages (P > 0.05). The same analyses were performed using Surveillance, Epidemiology, and End Results data, and all the observed results were validated.

CONCLUSION

The current AJCC T stage classification categorizes patients into groups with heterogenous prognosis, thus failing to serve as an effective staging tool. A modified T-stage classification demonstrated significantly improved stratification for patients with gastric neuroendocrine tumors.

A Robust Circular RNA-Associated Three-Gene Prognostic Signature for Patients with Gastric Cancer

Accumulating evidence has demonstrated that circular RNAs (circRNAs) play vital roles in cancer progression. However, the underlying molecular mechanisms of circRNAs remain poorly elucidated in gastric cancer (GC). The main purpose of present study is to explore the underlying regulatory mechanism by constructing a circRNA-associated competitive endogenous RNA (ceRNA) network and further establish a robust prognostic signature for patients with GC. Based on expression data of circRNA, microRNA, and mRNA derived from Gene Expression Omnibus (GEO) and The Cancer Genome Atlas (TCGA) databases, a circRNA-associated ceRNA network, containing 15 cirRNAs, 9 microRNAs, and 35 mRNAs, was constructed using the Starbase database. Functional enrichment analysis showed that the ceRNA network might be involved in many cancer-related pathways, such as regulation of transcription from RNA polymerase II promoter, mesodermal cell differentiation, and focal adhesion. A protein-protein interaction network was constructed based on genes within the circRNA-associated ceRNA network. We found that six of ten hub genes within the PPI network were significantly associated with overall survival (OS). Thus, using the LASSO method, we constructed a three-gene prognostic signature based on TCGA-GC cohort, which could classify GC patients into low-risk and high-risk groups with significant difference in OS (HR = 1.9, 95%CI = 1.14‐3.2, and log-rank p = 0.001). The prognostic performance of the three-gene signature was verified in GSE15459 (HR = 1.9, 95%CI = 1.27‐3.0, and log − rank p = 2.2E − 05) and GSE84437 (HR = 1.5, 95%CI = 1.17‐2.0, and log − rank p = 6.3E − 04). Multivariate Cox analysis further revealed that the three-gene prognostic signature could serve as an independent risk factor for OS. Taken together, our findings contribute to a better understanding of the underlying mechanisms of circRNAs in GC progression. Furthermore, a robust prognostic signature is meaningful to facilitate individualized treatment for patients with GC.

Immune gene prognostic signature for disease free survival of gastric cancer: Translational research of an artificial intelligence survival predictive system

The progress of artificial intelligence algorithms and massive data provide new ideas and choices for individual mortality risk prediction for cancer patients. The current research focused on depict immune gene related regulatory network and develop an artificial intelligence survival predictive system for disease free survival of gastric cancer.

Multi-task logistic regression algorithm, Cox survival regression algorithm, and Random survival forest algorithm were used to develop the artificial intelligence survival predictive system.

Nineteen transcription factors and seventy immune genes were identified to construct a transcription factor regulatory network of immune genes. Multivariate Cox regression identified fourteen immune genes as prognostic markers. These immune genes were used to construct a prognostic signature for gastric cancer. Concordance indexes were 0.800, 0.809, and 0.856 for 1-, 3- and 5- year survival. An interesting artificial intelligence survival predictive system was developed based on three artificial intelligence algorithms for gastric cancer. Gastric cancer patients with high risk score have poor survival than patients with low risk score.

The current study constructed a transcription factor regulatory network and developed two artificial intelligence survival prediction tools for disease free survival of gastric cancer patients. These artificial intelligence survival prediction tools are helpful for individualized treatment decision.

Junctional Adhesion Molecule-Like Protein Promotes Tumor Progression and Metastasis via p38 Signaling Pathway in Gastric Cancer

Junctional adhesion molecule-like protein (JAML), a newly discovered junctional adhesion molecule (JAM), mediates the adhesion and migration processes of various immune cells and endothelial/epithelial cells, ultimately regulating inflammation reaction. However, its role in tumors remains to be determined. The expression of JAML was examined in gastric cancer (GC) and peritumoral tissues from 63 patients. The relationship between JAML expression and clinical characteristics was also observed. In vitro, GC cell migration and proliferation were assessed by wound healing assay, transwell migration assay and EdU incorporation assay. Immunohistochemical staining results showed that JAML expression level was higher in GC tissues than in peritumoral tissues. High expression of JAML in cancer tissues was associated with worse cell differentiation, local lymph node involvement, deep infiltration, and advanced stage. In vitro, we found that JAML silencing inhibited GC cell migration and proliferation, while JAML overexpression promoted GC cell migration and proliferation, partially via p38 signaling. Taken together, our study revealed a critical role for JAML to promote GC cell migration and proliferation. JAML might be a novel diagnostic biomarker and therapeutic target for GC.

A risk score model with five long non-coding RNAs for predicting prognosis in gastric cancer: an integrated analysis combining TCGA and GEO datasets

Background

Gastric cancer (GC) is one of the most common carcinomas of the digestive tract, and the prognosis for these patients may be poor. There is evidence that some long non-coding RNAs(lncRNAs) can predict the prognosis of patients with GC. However, few lncRNA signatures have been used to predict prognosis. Herein, we aimed to construct a risk score model based on the expression of five lncRNAs to predict the prognosis of patients with GC and provide new potential therapeutic targets.

Methods

We performed differentially expressed and survival analyses to identify differentially expressed survival-ralated lncRNAs by using GC patient expression profile data from The Cancer Genome Atlas (TCGA) database. We then established a formula including five lncRNAs to predict the prognosis of patients with GC. In addition, to verify the prognostic value of this risk score model, two independent Gene Expression Omnibus (GEO) datasets, GSE62254 (N = 300) and GSE15459 (N = 200), were employed as validation groups.

Results

Based on the characteristics of five lncRNAs, patients with GC were divided into high or low risk subgroups. The prognostic value of the risk score model with five lncRNAs was confirmed in both TCGA and the two independent GEO datasets. Furthermore, stratification analysis results showed that this model had an independent prognostic value in patients with stage II-IV GC. We constructed a nomogram model combining clinical factors and the five lncRNAs to increase the accuracy of prognostic prediction. Enrichment analysis based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) suggested that the five lncRNAs are associated with multiple cancer occurrence and progression-related pathways.

Conclusion

The risk score model including five lncRNAs can predict the prognosis of patients with GC, especially those with stage II-IV, and may provide potential therapeutic targets in future.

The Edmonton Obesity Staging System Predicts Perioperative Complications and Procedure Choice in Obesity and Metabolic Surgery-a German Nationwide Register-Based Cohort Study (StuDoQ|MBE)

OBJECTIVE

To examine the relationship between Edmonton Obesity Staging System (EOSS) and perioperative complications as well as surgical procedure.

BACKGROUND

The application of EOSS for the selection of patients with obesity is a more comprehensive measure of obesity-related diseases and a predictor of mortality than body mass index (BMI).

METHODS

This was a nationwide cohort study using prospectively inserted data from the German register for obesity and metabolic surgery StuDoQ|MBE. All patients undergoing sleeve gastrectomy (SG), Roux-en Y gastric bypass (RYGB), and one-anastomosis gastric bypass (OAGB) between February 2015 and July 2017 as a primary treatment for severe obesity were included. Data included gender, age, BMI, ASA score, EOSS, early postoperative complications next to the Clavien-Dindo grading system, readmission, and 30-day mortality.

RESULTS

A total of 9437 patients were included. The mean BMI was 49.5 kg/m² ± 7.8 (range 35-103.5). The total postoperative complication rate was 5.3%, with the highest rate in EOSS 3 (7.8%) and 4 (6.8%). Thirty-day mortality was 0.2% with the highest mortality after SG in EOSS 3 (1.16%) and EOSS 4 (0.92%) (p = 0.0068). Crosstabs showed a prevalence of Clavien-Dindo III and IV complications of 3.4% (SG), 3.6% (RYGB), and 1.6% (OAGB) in EOSS 2 (p = 0.0032) and 3.5% (SG), 5.1% (RYGB), and 5.6% (OAGB) in EOSS 3.

CONCLUSION

The highest postoperative complications and mortality occurred in patients with EOSS ≥ 3. SG and OAGB could be the procedure of choice to reduce perioperative morbidity; nevertheless, it has to be in mind that in EOSS ≥ 3, SG has the highest mortality.

TRIAL REGISTRATION

ClinicalTrials.gov Identifier NCT03556059.

KIF15 facilitates gastric cancer via enhancing proliferation, inhibiting apoptosis, and predict poor prognosis

Background

Kinesin superfamily proteins (KIFs) can transport membranous organelles and protein complexes in an ATP-dependent manner. Kinesin family member 15 (KIF15) is overexpressed in various cancers. However, the function of KIF15 in gastric cancer (GC) is still unclear.

Methods

GC patients’ data from The Cancer Genome Atlas (TCGA) were analyzed by bioinformatics methods. The expression of KIF15 was examined in GC and paracarcinoma tissues from 41 patients to verify the analysis results. The relationship between KIF15 expression and clinical characteristics were also observed by bioinformatics methods. Kaplan–Meier survival analysis of 122 GC patients in our hospital was performed to explore the relationship between KIF15 expression levels and GC patients’ prognosis. KIF15 was downregulated in GC cell lines AGS and SGC-7901 by transfecting a lentivirus-mediated shRNA plasmid targeting KIF15. In vitro, GC cell proliferation and apoptosis were detected by MTT assay, colony formation assay, and Annexin V-APC staining. In vivo, xenograft experiments were used to verify the in vitro results. Furthermore, Human Apoptosis Antibody Array kit was used to screen possible targets of KIF15 in GC cell lines.

Results

The bioinformatics results showed that KIF15 expression levels were higher in GC tissues than in normal tissues. IHC showed same results. High expression of KIF15 was statistical correlated with high age and early histologic stage. Kaplan–Meier curves indicated that high KIF15 expression predict poor prognosis in patients with GC. MTT assay and colony formation assay showed that KIF15 promote GC cell proliferation. Annexin V-APC staining found that KIF15 can inhibit GC cell apoptosis. Xenograft experiments reveal that downregulating KIF15 can inhibit GC tumor growth and promote GC apoptosis. Through detection of 43 anti-apoptotic proteins by the Human Apoptosis Antibody Array kit, it was confirmed that knocking down KIF15 can reduce seven anti-apoptotic proteins expression.

Conclusions

Taken together, our study revealed a critical role for KIF15 to inhibit GC cell apoptosis and promote GC cell proliferation. KIF15 may decrease anti-apoptotic proteins expression by regulating apoptosis pathways. High expression of KIF15 predicts a poor prognosis in patients with GC. KIF15 might be a novel prognostic biomarker and a therapeutic target for GC.

MiR-652 serves as a prognostic biomarker in gastric cancer and promotes tumor proliferation, migration, and invasion via targeting RORA

PURPOSE

MicroRNAs (miRNAs) have been reported to be involved in tumorigenesis. The aim of this study was to investigate the functional role and prognostic value of miR-652 in gastric cancer (GC).

METHODS

Quantitative real-time polymerase chain reaction (qRT-PCR) was used to determine the expression levels of miR-652 in human GC tissue samples and GC cell lines. The Kaplan-Meier survival curves and Cox regression analysis were performed to measure the prognostic value of miR-652 in GC. The tumor cell proliferation capacity was estimated by MTT assay, and cell migration and invasion were assessed by Transwell assays. The luciferase reporter assay was performed to confirm the target gene of miR-652.

RESULTS

MiR-652 was significantly elevated in GC tissues and cell lines (all P< 0.001). And the expression level of miR-652 was significantly associated with TNM stage and lymph node metastasis (all P< 0.05). GC patients with high expression of miR-652 had a shorter overall survival rate than those with low miR-652 expression (log-rank P< 0.001). The miR-652 and TNM stage were proven to be independent prognostic predictors for the GC patients. Overexpressing miR-652 could enhance cell proliferation, migration and invasion (all P< 0.01). RORA was proved to be the target gene of miR-652.

CONCLUSION

MiR-652 functions as an oncogene in GC and promotes tumor progression via targeting RORA. MiR-652 might be a novel predictive marker for the poor prognosis of GC patients.

Genome-Wide Identification and Characterization of DNA Methylation and Long Non-Coding RNA Expression in Gastric Cancer

Abnormal DNA methylation, an epigenetic modification, has increasingly been linked to the pathogenesis of many human cancers. However, there has been little focus on the DNA methylation patterns of genes encoding long noncoding RNAs (lncRNAs) in gastric cancer (GC). This study comprehensively determined DNA methylation and lncRNA expression profiles in GC through genome-wide analysis. Differentially methylated loci and lncRNAs were identified by integrating multi-omics data. In total, 548 differentially methylated CpG sites in lncRNA promoters and 2,399 differentially expressed lncRNAs were screened that were capable of distinguishing GC from normal tissues. Among them, 22 differentially methylation sites in 17 lncRNAs were inversely related to expression levels. Further analysis of DNA methylation status and gene expression level in GC revealed that three CpG sites (cg01550148, cg22497867, and cg20001829) and two lncRNAs (RP11-366F6.2 and RP5-881L22.5) were significantly associated with GC patient overall survival. Molecular function analysis showed that these abnormally methylated lncRNAs were mainly involved in transcriptional activator activity. Our study identified several lncRNAs regulated by aberrant DNA methylation that have clinical utility as novel prognostic biomarkers in GC. These findings help improve the understanding of methylated patterns of lncRNAs and further our knowledge of the role of epigenetics in cancer development.

Long non-coding RNA ERICH3-AS1 is an unfavorable prognostic factor for gastric cancer

Long non-coding RNAs (lncRNAs) play important roles in gastric cancer (GC), but the mechanism is not fully clear. ERICH3-AS1 (ERICH3 antisense RNA1) is affiliated with the non-coding RNA class which has proven to be involved in the prognostic of GC, but the function of ERICH3-AS1 is still unclear. In this study, we aim to explore the potential function of ERICH3-AS1 in the development of GC and analyze the prognostic role of ERICH3-AS1 in GC. We found that the lncRNA ERICH3-AS1 was significantly up-regulated in GC tissues in the analysis of The Cancer Genome Atlas (TCGA) data; the Kaplan-Meier analysis showed that the higher the expression of ERICH3-AS1 was, the earlier the recurrence and the poorer the prognosis would be in patients. Cox univariate and multivariate analyses revealed that ERICH3-AS1 was a risk factor of disease-free survival (DFS) (p < 0.05) and overall survival (OS) (p < 0.05) of patients. Through Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses, it demonstrated that the ERBB pathways, the mitogen-activated protein kinase (MAPK) pathways, the MTOR pathways, p53 pathways and Wnt pathways were differentially enriched in ERICH3-AS1 high expression phenotype. Furthermore, the correlation analysis showed that ERICH3-AS1 had significant correlations with apoptosis-related proteins such as BCL2L10 and CASP14; cell cycle-associated proteins CDK14 and invasion and migration-associated proteins such as MMP20, MMP26 and MMP27. In summary, we identified that increased ERICH3-AS1 might be a potential biomarker for diagnosis and independent prognostic factor of GC. Moreover, ERICH3-AS1 might participate in the oncogenesis and development of tumors via cell cycle and apoptosis pathway mediated by ERBB, MAPK, MTOR, p53 and Wnt pathways.

The regulatory roles and potential prognosis implications of long non-coding RNAs in gastric cancer

Accumulating dysregulated lncRNAs have been demonstrated to execute vital functions in the pathogenesis and progress of gastric cancer (GC) through versatile molecular mechanisms. In this review, we classify the mechanisms of dysregulated lncRNAs in GC into several governing types according to their roles at molecular level. For each regulatory role, we illustrate several instructive examples and introduce significant effects of lncRNAs on cellular biological properties of GC. Besides, we summarize a group of lncRNA-signatures that are potential biomarkers in the prediction of prognosis for GC patients.

Integrated analysis reveals potential long non-coding RNA biomarkers and their potential biological functions for disease free survival in gastric cancer patients

Background

Increasing evidences supported the association between long non-coding RNA (lncRNA) and disease free survival in gastric cancer (GC) patients. The purpose of the current study was to construct and verify a noninvasive preoperative predictive tool for disease free survival in GC patients.

Methods

There were 265 and 300 GC patients in model dataset and validation dataset respectively. The associations between the lncRNA biomarkers and disease free survival were evaluated by univariate and multivariate Cox regression.

Results

Thirteen lncRNA biomarkers (GAS5-AS1, AL109615.3, KDM7A-DT, AP000866.2, KCNJ2-AS1, LINC00656, LINC01777, AC046185.3, TTTY14, LINC01526, LINC02523, LINC00592, and C5orf66) were identified as prognostic biomarkers with disease free survival. These thirteen lncRNA biomarkers were combined to construct a prognostic signature for disease free survival. The C-indexes of the current predictive signature in model cohort were 0.849 (95% CI 0.803–0.895), 0.859 (95% CI 0.813–0.905) and 0.888 (95% CI 0.842–0.934) for 1-year, 3-year and 5-year disease free survival respectively. Based on thirteen-lncRNA prognostic signature, patients in model cohort could be stratified into high risk group and low risk group with significant different disease free survival rate (hazard ratio [HR] = 7.355, 95% confidence interval [CI] 4.378–12.356). Good reproducibility of thirteen-lncRNA prognostic signature was confirmed in an external validation cohort (GSE62254) with HR 3.919 and 95% CI 2.817–5.453. Further analysis demonstrated that the prognostic significance of thirteen-lncRNA prognostic signature was independent of other clinical characteristics.

Conclusions

In conclusion, a simple noninvasive prognostic signature was established for preoperative prediction of disease free survival in GC patients. This prognostic signature might predict the individual mortality risk of disease free survival without pathological information and facilitate individual treatment decision-making.

Electronic supplementary material

The online version of this article (10.1186/s12935-019-0846-6) contains supplementary material, which is available to authorized users.